A model = repCleaner + featureGenerator + featureProcessor + MedPredictor. More...
#include <MedModel.h>
Inheritance diagram for MedModel:
Public Member Functions | |
| void | clear () |
| void | change_model (const ChangeModelInfo &change_request) |
| change model object in run time | |
| void | change_model (const vector< ChangeModelInfo > &change_request) |
| change model object in run time - multiple requests, one by one | |
| void | add_rep_processor (RepProcessor *processor) |
| Add Rep Processorsep. | |
| void | add_rep_processors_set (RepProcessorTypes type, vector< string > &signals) |
| void | add_rep_processors_set (RepProcessorTypes type, vector< string > &signals, string init_string) |
| void | insert_rep_processor (string init_string, int index) |
| void | add_feature_generator (FeatureGenerator *generator) |
| Add Feature Generators. | |
| void | add_feature_generators (FeatureGeneratorTypes type, vector< string > &signals) |
| void | add_feature_generators (FeatureGeneratorTypes type, vector< string > &signals, string init_string) |
| void | add_feature_generator (FeatureGeneratorTypes type, string &signal) |
| void | add_feature_generators (FeatureGeneratorTypes type, string &signal, string init_string) |
| void | add_feature_generators (string &name, vector< string > &signals) |
| void | add_feature_generators (string &name, vector< string > &signals, string init_string) |
| void | add_feature_generator (string &name, string &signal) |
| void | add_feature_generators (string &name, string &signal, string init_string) |
| void | add_age () |
| void | add_gender () |
| void | get_all_features_names (vector< string > &feat_names, int before_process_set) |
| void | add_feature_processor (FeatureProcessor *processor) |
| Add Feature Processors. | |
| void | add_feature_processors_set (FeatureProcessorTypes type) |
| void | add_feature_processors_set (FeatureProcessorTypes type, string init_string) |
| void | add_feature_processors_set (FeatureProcessorTypes type, vector< string > &features) |
| void | add_feature_processors_set (FeatureProcessorTypes type, vector< string > &features, string init_string) |
| void | add_normalizers () |
| void | add_normalizers (string init_string) |
| void | add_normalizers (vector< string > &features) |
| void | add_normalizers (vector< string > &features, string init_string) |
| void | add_imputers () |
| void | add_imputers (string init_string) |
| void | add_imputers (vector< string > &features) |
| void | add_imputers (vector< string > &features, string init_string) |
| void | add_post_processor (PostProcessor *processor) |
| int | init_from_json_string (string &json_string, const string &fname) |
| void | init_from_json_file (const string &fname) |
| void | init_from_json_file_with_alterations_version_1 (const string &fname, vector< string > &alterations) |
| void | init_from_json_file_with_alterations (const string &fname, vector< string > &alterations) |
| int | add_pre_processors_json_string_to_model (string in_json, string fname) |
| int | add_pre_processors_json_string_to_model (string in_json, string fname, vector< string > &alterations, bool add_rep_first=false) |
| int | add_post_processors_json_string_to_model (string in_json, string fname) |
| int | add_post_processors_json_string_to_model (string in_json, string fname, vector< string > &alterations) |
| void | add_rep_processor_to_set (int i_set, const string &init_string) |
| void | add_feature_generator_to_set (int i_set, const string &init_string) |
| void | add_feature_processor_to_set (int i_set, int duplicate, const string &init_string) |
| void | add_process_to_set (int i_set, int duplicate, const string &init_string) |
| void | add_process_to_set (int i_set, const string &init_string) |
| void | add_post_processor_to_set (int i_set, const string &init_string) |
| void | set_predictor (MedPredictor *_predictor) |
| Add Predictor. | |
| void | set_predictor (MedPredictorTypes type) |
| void | set_predictor (string name) |
| void | set_predictor (MedPredictorTypes type, string init_string) |
| void | set_predictor (string name, string init_string) |
| void | replace_predictor_with_json_predictor (string f_json) |
| void | set_required_signal_ids (MedDictionarySections &dict, vector< RepProcessor * > &applied_rep_processors, vector< FeatureGenerator * > &applied_generators) |
| signal ids | |
| void | set_affected_signal_ids (MedDictionarySections &dict, vector< RepProcessor * > &applied_rep_processors) |
| void | get_required_signal_names (unordered_set< string > &signalNames) const |
| void | get_required_signal_names (vector< string > &signalNames) const |
| void | get_required_signal_names_for_processed_values (unordered_set< string > &targetSignalNames, unordered_set< string > &signalNames) |
| Required signals to generate processed values of target-signals. | |
| void | get_required_signal_names_for_processed_values (unordered_set< string > &targetSignalNames, vector< string > &signalNames) |
| void | get_generated_features_names (vector< string > &feat_names) |
| Get list of Features generated by the model, after everything was applied (RPs, FGs, FPs) | |
| int | get_nfeatures () |
| get model features size esstimation - number of columns | |
| int | get_duplicate_factor () const |
| returns the duplicate factor of the model - by how much each sample is duplicated by | |
| int | collect_and_add_virtual_signals (MedRepository &rep) |
| void | get_required_signal_categories (unordered_map< string, vector< string > > &signal_categories_in_use) const |
| void | init_all (MedDictionarySections &dict, MedSignals &sigs, vector< RepProcessor * > &applied_rep_processors, vector< FeatureGenerator * > &applied_generators) |
| Initialization : signal ids and tables. | |
| void | init_all (MedDictionarySections &dict, MedSignals &sigs) |
| int | learn (MedPidRepository &rep, MedSamples *samples) |
| int | learn (MedPidRepository &rep, MedSamples *samples, MedModelStage start_stage, MedModelStage end_stage) |
| int | learn_skip_matrix_train (MedPidRepository &rep, MedSamples *samples, MedModelStage end_stage) |
| int | apply (MedPidRepository &rep, MedSamples &samples) |
| int | apply (MedPidRepository &rep, MedSamples &samples, MedModelStage start_stage, MedModelStage end_stage) |
| void | no_init_apply_partial (MedPidRepository &rep, MedSamples &samples, const vector< Effected_Field > &requested_outputs) |
| int | init_model_for_apply (MedPidRepository &rep, MedModelStage start_stage, MedModelStage end_stage) |
| int | no_init_apply (MedPidRepository &rep, MedSamples &samples, MedModelStage start_stage, MedModelStage end_stage) |
| int | learn (MedPidRepository &rep, MedSamples &model_learning_set, vector< MedSamples > &post_processors_learning_sets) |
| int | learn (MedPidRepository &rep, MedSamples &model_learning_set, vector< MedSamples > &post_processors_learning_sets, MedModelStage start_stage, MedModelStage end_stage) |
| int | learn (MedPidRepository &rep, MedSamples &samples) |
| int | learn (MedPidRepository &rep, MedSamples &samples, MedModelStage start_stage, MedModelStage end_stage) |
| int | init_for_apply_rec (MedPidRepository &rep) |
| int | apply_rec (PidDynamicRec &drec, MedIdSamples idSamples, MedFeatures &_feat, bool copy_rec_flag, int end_stage=MED_MDL_APPLY_FTR_PROCESSORS) |
| virtual void | pre_serialization () |
| int | quick_learn_rep_processors (MedPidRepository &rep, MedSamples &samples) |
| int | learn_rep_processors (MedPidRepository &rep, MedSamples &samples) |
| int | learn_all_rep_processors (MedPidRepository &rep, MedSamples &samples) |
| void | filter_rep_processors () |
| int | learn_feature_generators (MedPidRepository &rep, MedSamples *learn_samples) |
| int | generate_features (MedPidRepository &rep, MedSamples *samples, vector< FeatureGenerator * > &_generators, MedFeatures &features) |
| int | generate_all_features (MedPidRepository &rep, MedSamples *samples, MedFeatures &features, unordered_set< string > &req_feature_generators) |
| int | learn_and_apply_feature_processors (MedFeatures &features) |
| int | learn_feature_processors (MedFeatures &features) |
| int | apply_feature_processors (MedFeatures &features, bool learning) |
| int | apply_feature_processors (MedFeatures &features, vector< unordered_set< string > > &req_features_vec, bool learning) |
| void | build_req_features_vec (vector< unordered_set< string > > &req_features_vec) const |
| void | get_applied_generators (unordered_set< string > &req_feature_generators, vector< FeatureGenerator * > &_generators) const |
| void | dprint_process (const string &pref, int rp_flag, int fg_flag, int fp_flag, int predictor_flag, int pp_flag) |
| following is for debugging, it gets a prefix, and prints it along with information on rep_processors, feature_generators, or feature_processors | |
| int | write_feature_matrix (const string mat_fname, bool write_attributes=false, bool append=false) |
| following is for debugging : writing the feature to a csv file as a matrix. | |
| void | load_repository (const string &configFile, MedPidRepository &rep, bool allow_adjustment=false) |
| loading a repository (optionally allowing for adjustment to model according to available signals) | |
| void | load_repository (const string &configFile, vector< int > ids, MedPidRepository &rep, bool allow_adjustment=false) |
| void | fit_for_repository (MedPidRepository &rep) |
| void | read_from_file_with_changes (const string &model_binary_path, const string &path_to_json_changes) |
| Read binary model from file + json changes req for run-time (empty string for no changes) | |
| void | clone_model (MedModel &out) |
| clones this object into out | |
| void | copy_from_model (MedModel &in) |
| copy in modle into this object | |
| int | get_apply_batch_count () |
| returns how many samples are done in a single batche (in apply) | |
 Public Member Functions inherited from SerializableObject | |
| virtual int | version () const |
| Relevant for serializations. | |
| virtual string | my_class_name () const |
| For better handling of serializations it is highly recommended that each SerializableObject inheriting class will implement the next method. | |
| virtual void | serialized_fields_name (vector< string > &field_names) const |
| The names of the serialized fields. | |
| virtual void * | new_polymorphic (string derived_name) |
| for polymorphic classes that want to be able to serialize/deserialize a pointer * to the derived class given its type one needs to implement this function to return a new to the derived class given its type (as in my_type) | |
| virtual void | post_deserialization () |
| virtual size_t | get_size () |
| Gets bytes sizes for serializations. | |
| virtual size_t | serialize (unsigned char *blob) |
| Serialiazing object to blob memory. return number ob bytes wrote to memory. | |
| virtual size_t | deserialize (unsigned char *blob) |
| Deserialiazing blob to object. returns number of bytes read. | |
| size_t | serialize_vec (vector< unsigned char > &blob) |
| size_t | deserialize_vec (vector< unsigned char > &blob) |
| virtual size_t | serialize (vector< unsigned char > &blob) |
| virtual size_t | deserialize (vector< unsigned char > &blob) |
| virtual int | read_from_file (const string &fname) |
| read and deserialize model | |
| virtual int | write_to_file (const string &fname) |
| serialize model and write to file | |
| virtual int | read_from_file_unsafe (const string &fname) |
| read and deserialize model without checking version number - unsafe read | |
| int | init_from_string (string init_string) |
| Init from string. | |
| int | init_params_from_file (string init_file) |
| int | init_param_from_file (string file_str, string ¶m) |
| virtual int | init (map< string, string > &map) |
| Virtual to init object from parsed fields. | |
| int | update_from_string (const string &init_string) |
| virtual int | update (map< string, string > &map) |
| Virtual to update object from parsed fields. | |
| virtual string | object_json () const |
Static Public Member Functions | |
| static MedModelStage | get_med_model_stage (const string &stage) |
Data Fields | |
| string | version_info = "" |
| a string that represents version info - filled in compile time from git info | |
| int | serialize_learning_set = 0 |
| remember learning set | |
| int | model_json_version = 1 |
| the json version | |
| long long | max_data_in_mem = 0 |
| maximal data size in memory. | |
| int | model_feature_count_hint = 0 |
| hint for feature counts to consider in batching. | |
| vector< RepProcessor * > | rep_processors |
| Repostiroy-level cleaners; to be applied sequentially. | |
| vector< FeatureGenerator * > | generators |
| Feature Generators. | |
| vector< FeatureProcessor * > | feature_processors |
| Features-level cleaners; to be applied sequentially. | |
| vector< PostProcessor * > | post_processors |
| Post Process level - calibrators, explainers... | |
| MedPredictor * | predictor = NULL |
| Predictor. | |
| MedSamples * | LearningSet = NULL |
| Learning samples. | |
| bool | take_mean_pred = true |
| when having multiple prediction for same samples - how to aggregate preds - mean or median? | |
| unordered_set< string > | required_signal_names |
| All required signal names + ids. | |
| unordered_set< int > | required_signal_ids |
| vector< unordered_set< string > > | required_features_vec |
| unordered_set< string > | required_feature_generators |
| map< string, int > | virtual_signals |
| all collected virtual signals (name to type) | |
| map< string, string > | virtual_signals_generic |
| MedFeatures | features |
| the created matrix - no need to serialize | |
| int | verbosity = 1 |
| verbosity 0 -> much less printouts in predict | |
| int | generate_masks_for_features = 0 |
| MedPidRepository * | p_rep = NULL |
| not serialized. stores pointer to rep used in Learn or Apply after call. | |
Detailed Description
A model = repCleaner + featureGenerator + featureProcessor + MedPredictor.
Member Function Documentation
◆ change_model() [1/2]
| void MedModel::change_model | ( | const ChangeModelInfo & | change_request | ) |
change model object in run time
- Parameters
-
change_request.object_type_name - object type name to search for in rep_processors,generators,feature_processors, etc. For example "FeatureNormalizer" change_request.json_query query on the object json to filter on specific attributes. leave empty to operate on all change_request.change_command - The command to send each matched object. use "DELETE" to remove the object, "PRINT" to print object josn into stdout. otherwise it will pass the argument into init function
◆ change_model() [2/2]
| void MedModel::change_model | ( | const vector< ChangeModelInfo > & | change_request | ) |
change model object in run time - multiple requests, one by one
- Parameters
-
change_request.object_type_name - object type name to search for in rep_processors,generators,feature_processors, etc. For example "FeatureNormalizer" change_request.json_query query on the object json to filter on specific attributes. leave empty to operate on all change_request.change_command - The command to send each matched object. use "DELETE" to remove the object, "PRINT" to print object josn into stdout. otherwise it will pass the argument into init function
◆ pre_serialization()
|
inlinevirtual |
Reimplemented from SerializableObject.
Field Documentation
◆ max_data_in_mem
| long long MedModel::max_data_in_mem = 0 |
maximal data size in memory.
when <=0 mean auto mode
◆ model_feature_count_hint
| int MedModel::model_feature_count_hint = 0 |
hint for feature counts to consider in batching.
The user can specify count for model after feature selection, otherwise it will count all the features from the generators without filtering. The number of features needs to be defined pre ahead of batching process of the samples.
The documentation for this class was generated from the following files:
- Internal/MedProcessTools/MedProcessTools/MedModel.h
- Internal/MedProcessTools/MedProcessTools/MedModel.cpp
- Internal/MedProcessTools/MedProcessTools/MedModelInitFromJson.cpp
